{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9973216459702945,
  "eval_steps": 100,
  "global_step": 1026,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009739469198928659,
      "grad_norm": 5.16649192443276,
      "learning_rate": 4.854368932038835e-07,
      "loss": 0.956,
      "mean_token_accuracy": 0.7770209729671478,
      "step": 5
    },
    {
      "epoch": 0.019478938397857318,
      "grad_norm": 4.487616874759018,
      "learning_rate": 9.70873786407767e-07,
      "loss": 0.9595,
      "mean_token_accuracy": 0.774706457555294,
      "step": 10
    },
    {
      "epoch": 0.029218407596785977,
      "grad_norm": 1.9918521253065147,
      "learning_rate": 1.4563106796116506e-06,
      "loss": 0.9087,
      "mean_token_accuracy": 0.7794785097241401,
      "step": 15
    },
    {
      "epoch": 0.038957876795714635,
      "grad_norm": 2.140741610293528,
      "learning_rate": 1.941747572815534e-06,
      "loss": 0.8516,
      "mean_token_accuracy": 0.7845764443278312,
      "step": 20
    },
    {
      "epoch": 0.048697345994643294,
      "grad_norm": 1.577104941303547,
      "learning_rate": 2.427184466019418e-06,
      "loss": 0.7835,
      "mean_token_accuracy": 0.795551997423172,
      "step": 25
    },
    {
      "epoch": 0.05843681519357195,
      "grad_norm": 0.9851513931833208,
      "learning_rate": 2.912621359223301e-06,
      "loss": 0.7536,
      "mean_token_accuracy": 0.801626966893673,
      "step": 30
    },
    {
      "epoch": 0.0681762843925006,
      "grad_norm": 0.7665830684245213,
      "learning_rate": 3.398058252427185e-06,
      "loss": 0.7207,
      "mean_token_accuracy": 0.8073863789439202,
      "step": 35
    },
    {
      "epoch": 0.07791575359142927,
      "grad_norm": 0.6623786102965908,
      "learning_rate": 3.883495145631068e-06,
      "loss": 0.6976,
      "mean_token_accuracy": 0.8124146014451981,
      "step": 40
    },
    {
      "epoch": 0.08765522279035792,
      "grad_norm": 0.5409465013517523,
      "learning_rate": 4.368932038834952e-06,
      "loss": 0.6791,
      "mean_token_accuracy": 0.8153241157531739,
      "step": 45
    },
    {
      "epoch": 0.09739469198928659,
      "grad_norm": 0.533911697336328,
      "learning_rate": 4.854368932038836e-06,
      "loss": 0.6494,
      "mean_token_accuracy": 0.8212776482105255,
      "step": 50
    },
    {
      "epoch": 0.10713416118821524,
      "grad_norm": 0.46335668910229383,
      "learning_rate": 5.3398058252427185e-06,
      "loss": 0.6374,
      "mean_token_accuracy": 0.8242872759699822,
      "step": 55
    },
    {
      "epoch": 0.1168736303871439,
      "grad_norm": 0.48547854119926936,
      "learning_rate": 5.825242718446602e-06,
      "loss": 0.6321,
      "mean_token_accuracy": 0.8252887204289436,
      "step": 60
    },
    {
      "epoch": 0.12661309958607256,
      "grad_norm": 0.4864652293922326,
      "learning_rate": 6.310679611650487e-06,
      "loss": 0.6246,
      "mean_token_accuracy": 0.8261281028389931,
      "step": 65
    },
    {
      "epoch": 0.1363525687850012,
      "grad_norm": 0.48690138801181443,
      "learning_rate": 6.79611650485437e-06,
      "loss": 0.6168,
      "mean_token_accuracy": 0.8278339207172394,
      "step": 70
    },
    {
      "epoch": 0.14609203798392986,
      "grad_norm": 0.4337713878880355,
      "learning_rate": 7.2815533980582534e-06,
      "loss": 0.5925,
      "mean_token_accuracy": 0.8333460614085197,
      "step": 75
    },
    {
      "epoch": 0.15583150718285854,
      "grad_norm": 0.45660620623082415,
      "learning_rate": 7.766990291262136e-06,
      "loss": 0.5973,
      "mean_token_accuracy": 0.8320671111345291,
      "step": 80
    },
    {
      "epoch": 0.1655709763817872,
      "grad_norm": 0.4424374164027964,
      "learning_rate": 8.25242718446602e-06,
      "loss": 0.5873,
      "mean_token_accuracy": 0.8338468298316002,
      "step": 85
    },
    {
      "epoch": 0.17531044558071585,
      "grad_norm": 0.4949541791856808,
      "learning_rate": 8.737864077669904e-06,
      "loss": 0.5846,
      "mean_token_accuracy": 0.8346328064799309,
      "step": 90
    },
    {
      "epoch": 0.1850499147796445,
      "grad_norm": 0.44716595203588544,
      "learning_rate": 9.223300970873788e-06,
      "loss": 0.574,
      "mean_token_accuracy": 0.8365551233291626,
      "step": 95
    },
    {
      "epoch": 0.19478938397857318,
      "grad_norm": 0.40922461158607365,
      "learning_rate": 9.708737864077671e-06,
      "loss": 0.5745,
      "mean_token_accuracy": 0.8366570115089417,
      "step": 100
    },
    {
      "epoch": 0.20452885317750183,
      "grad_norm": 0.5016773528604923,
      "learning_rate": 9.99988415036596e-06,
      "loss": 0.5649,
      "mean_token_accuracy": 0.8382768034934998,
      "step": 105
    },
    {
      "epoch": 0.21426832237643048,
      "grad_norm": 0.5010899629730204,
      "learning_rate": 9.99858090363555e-06,
      "loss": 0.569,
      "mean_token_accuracy": 0.8377967774868011,
      "step": 110
    },
    {
      "epoch": 0.22400779157535913,
      "grad_norm": 0.5010895934649102,
      "learning_rate": 9.995829976834402e-06,
      "loss": 0.5654,
      "mean_token_accuracy": 0.8383171066641808,
      "step": 115
    },
    {
      "epoch": 0.2337472607742878,
      "grad_norm": 0.5864787846174829,
      "learning_rate": 9.99163216668102e-06,
      "loss": 0.5703,
      "mean_token_accuracy": 0.8369988009333611,
      "step": 120
    },
    {
      "epoch": 0.24348672997321646,
      "grad_norm": 0.5502383423638442,
      "learning_rate": 9.985988688937684e-06,
      "loss": 0.5632,
      "mean_token_accuracy": 0.8387595832347869,
      "step": 125
    },
    {
      "epoch": 0.2532261991721451,
      "grad_norm": 0.4838385873501471,
      "learning_rate": 9.978901178058333e-06,
      "loss": 0.5472,
      "mean_token_accuracy": 0.8424718379974365,
      "step": 130
    },
    {
      "epoch": 0.26296566837107377,
      "grad_norm": 0.5371306966614764,
      "learning_rate": 9.970371686715205e-06,
      "loss": 0.5431,
      "mean_token_accuracy": 0.8432716697454452,
      "step": 135
    },
    {
      "epoch": 0.2727051375700024,
      "grad_norm": 0.5584506793424302,
      "learning_rate": 9.960402685204347e-06,
      "loss": 0.5516,
      "mean_token_accuracy": 0.8408507108688354,
      "step": 140
    },
    {
      "epoch": 0.2824446067689311,
      "grad_norm": 0.5467853450677332,
      "learning_rate": 9.948997060730161e-06,
      "loss": 0.5464,
      "mean_token_accuracy": 0.8424971371889114,
      "step": 145
    },
    {
      "epoch": 0.2921840759678597,
      "grad_norm": 0.48818575924287444,
      "learning_rate": 9.936158116569231e-06,
      "loss": 0.5489,
      "mean_token_accuracy": 0.8420207649469376,
      "step": 150
    },
    {
      "epoch": 0.30192354516678843,
      "grad_norm": 0.46723378629583207,
      "learning_rate": 9.921889571113629e-06,
      "loss": 0.5462,
      "mean_token_accuracy": 0.8419363871216774,
      "step": 155
    },
    {
      "epoch": 0.3116630143657171,
      "grad_norm": 0.561719955013033,
      "learning_rate": 9.906195556793996e-06,
      "loss": 0.546,
      "mean_token_accuracy": 0.8423524782061577,
      "step": 160
    },
    {
      "epoch": 0.32140248356464574,
      "grad_norm": 0.625160714021079,
      "learning_rate": 9.889080618882719e-06,
      "loss": 0.5335,
      "mean_token_accuracy": 0.8451830595731735,
      "step": 165
    },
    {
      "epoch": 0.3311419527635744,
      "grad_norm": 0.5878540451317104,
      "learning_rate": 9.870549714177538e-06,
      "loss": 0.5463,
      "mean_token_accuracy": 0.8417419150471688,
      "step": 170
    },
    {
      "epoch": 0.34088142196250304,
      "grad_norm": 0.6154889939045929,
      "learning_rate": 9.850608209565967e-06,
      "loss": 0.5327,
      "mean_token_accuracy": 0.8450453072786331,
      "step": 175
    },
    {
      "epoch": 0.3506208911614317,
      "grad_norm": 0.4329465610848603,
      "learning_rate": 9.829261880470941e-06,
      "loss": 0.5392,
      "mean_token_accuracy": 0.8434989348053932,
      "step": 180
    },
    {
      "epoch": 0.36036036036036034,
      "grad_norm": 0.5971707984256834,
      "learning_rate": 9.806516909178161e-06,
      "loss": 0.5324,
      "mean_token_accuracy": 0.84499292075634,
      "step": 185
    },
    {
      "epoch": 0.370099829559289,
      "grad_norm": 0.44190322710346147,
      "learning_rate": 9.78237988304557e-06,
      "loss": 0.5332,
      "mean_token_accuracy": 0.845010556280613,
      "step": 190
    },
    {
      "epoch": 0.3798392987582177,
      "grad_norm": 0.4258724885095506,
      "learning_rate": 9.756857792595555e-06,
      "loss": 0.5319,
      "mean_token_accuracy": 0.845000034570694,
      "step": 195
    },
    {
      "epoch": 0.38957876795714635,
      "grad_norm": 0.4729261486427116,
      "learning_rate": 9.729958029490353e-06,
      "loss": 0.5336,
      "mean_token_accuracy": 0.8447421163320541,
      "step": 200
    },
    {
      "epoch": 0.399318237156075,
      "grad_norm": 0.4528083565277609,
      "learning_rate": 9.701688384391296e-06,
      "loss": 0.5347,
      "mean_token_accuracy": 0.8443498685956001,
      "step": 205
    },
    {
      "epoch": 0.40905770635500366,
      "grad_norm": 0.5259149339810829,
      "learning_rate": 9.672057044702492e-06,
      "loss": 0.5199,
      "mean_token_accuracy": 0.848120279610157,
      "step": 210
    },
    {
      "epoch": 0.4187971755539323,
      "grad_norm": 0.46836949308672154,
      "learning_rate": 9.641072592199599e-06,
      "loss": 0.5219,
      "mean_token_accuracy": 0.8473641723394394,
      "step": 215
    },
    {
      "epoch": 0.42853664475286096,
      "grad_norm": 0.5302502163183013,
      "learning_rate": 9.608744000544392e-06,
      "loss": 0.5174,
      "mean_token_accuracy": 0.8485160410404206,
      "step": 220
    },
    {
      "epoch": 0.4382761139517896,
      "grad_norm": 0.51343914785562,
      "learning_rate": 9.575080632685832e-06,
      "loss": 0.5239,
      "mean_token_accuracy": 0.846913392841816,
      "step": 225
    },
    {
      "epoch": 0.44801558315071827,
      "grad_norm": 0.5925559680424145,
      "learning_rate": 9.54009223814837e-06,
      "loss": 0.5277,
      "mean_token_accuracy": 0.8454070091247559,
      "step": 230
    },
    {
      "epoch": 0.4577550523496469,
      "grad_norm": 0.426110283810703,
      "learning_rate": 9.503788950208324e-06,
      "loss": 0.5215,
      "mean_token_accuracy": 0.8473676040768623,
      "step": 235
    },
    {
      "epoch": 0.4674945215485756,
      "grad_norm": 0.5303378941560718,
      "learning_rate": 9.466181282959083e-06,
      "loss": 0.5282,
      "mean_token_accuracy": 0.8465007901191711,
      "step": 240
    },
    {
      "epoch": 0.4772339907475043,
      "grad_norm": 0.4965315113021109,
      "learning_rate": 9.427280128266049e-06,
      "loss": 0.5179,
      "mean_token_accuracy": 0.8484022691845894,
      "step": 245
    },
    {
      "epoch": 0.48697345994643293,
      "grad_norm": 0.44982624201366644,
      "learning_rate": 9.387096752612144e-06,
      "loss": 0.5224,
      "mean_token_accuracy": 0.8463509559631348,
      "step": 250
    },
    {
      "epoch": 0.4967129291453616,
      "grad_norm": 0.4623459721033935,
      "learning_rate": 9.345642793834825e-06,
      "loss": 0.5271,
      "mean_token_accuracy": 0.8463737353682518,
      "step": 255
    },
    {
      "epoch": 0.5064523983442902,
      "grad_norm": 0.5825923961149249,
      "learning_rate": 9.302930257755579e-06,
      "loss": 0.53,
      "mean_token_accuracy": 0.8450088694691658,
      "step": 260
    },
    {
      "epoch": 0.5161918675432189,
      "grad_norm": 0.45940800329264964,
      "learning_rate": 9.258971514702789e-06,
      "loss": 0.507,
      "mean_token_accuracy": 0.8508246764540672,
      "step": 265
    },
    {
      "epoch": 0.5259313367421475,
      "grad_norm": 0.4465364801256697,
      "learning_rate": 9.213779295929082e-06,
      "loss": 0.5087,
      "mean_token_accuracy": 0.8500014141201973,
      "step": 270
    },
    {
      "epoch": 0.5356708059410762,
      "grad_norm": 0.4686695606038613,
      "learning_rate": 9.167366689924116e-06,
      "loss": 0.5163,
      "mean_token_accuracy": 0.8484609499573708,
      "step": 275
    },
    {
      "epoch": 0.5454102751400048,
      "grad_norm": 0.5572145736502692,
      "learning_rate": 9.119747138623925e-06,
      "loss": 0.5221,
      "mean_token_accuracy": 0.8470426678657532,
      "step": 280
    },
    {
      "epoch": 0.5551497443389335,
      "grad_norm": 0.4765827682250088,
      "learning_rate": 9.070934433517872e-06,
      "loss": 0.5068,
      "mean_token_accuracy": 0.8509115263819694,
      "step": 285
    },
    {
      "epoch": 0.5648892135378621,
      "grad_norm": 0.49934327207360063,
      "learning_rate": 9.020942711654404e-06,
      "loss": 0.5106,
      "mean_token_accuracy": 0.8498208403587342,
      "step": 290
    },
    {
      "epoch": 0.5746286827367908,
      "grad_norm": 0.5101324493724546,
      "learning_rate": 8.969786451546691e-06,
      "loss": 0.5123,
      "mean_token_accuracy": 0.8496938437223435,
      "step": 295
    },
    {
      "epoch": 0.5843681519357194,
      "grad_norm": 0.5089021103596151,
      "learning_rate": 8.917480468979387e-06,
      "loss": 0.5128,
      "mean_token_accuracy": 0.8487787261605263,
      "step": 300
    },
    {
      "epoch": 0.5941076211346482,
      "grad_norm": 0.5501775930667125,
      "learning_rate": 8.864039912717713e-06,
      "loss": 0.5123,
      "mean_token_accuracy": 0.849444879591465,
      "step": 305
    },
    {
      "epoch": 0.6038470903335769,
      "grad_norm": 0.41563520519489466,
      "learning_rate": 8.809480260120096e-06,
      "loss": 0.5048,
      "mean_token_accuracy": 0.8513683333992959,
      "step": 310
    },
    {
      "epoch": 0.6135865595325055,
      "grad_norm": 0.4999791105897695,
      "learning_rate": 8.753817312655642e-06,
      "loss": 0.514,
      "mean_token_accuracy": 0.8484693005681038,
      "step": 315
    },
    {
      "epoch": 0.6233260287314342,
      "grad_norm": 0.48088239973601826,
      "learning_rate": 8.697067191327748e-06,
      "loss": 0.5114,
      "mean_token_accuracy": 0.8495015501976013,
      "step": 320
    },
    {
      "epoch": 0.6330654979303628,
      "grad_norm": 0.4440141428026334,
      "learning_rate": 8.639246332005163e-06,
      "loss": 0.5064,
      "mean_token_accuracy": 0.8507678374648094,
      "step": 325
    },
    {
      "epoch": 0.6428049671292915,
      "grad_norm": 0.5288324873519791,
      "learning_rate": 8.580371480661857e-06,
      "loss": 0.5024,
      "mean_token_accuracy": 0.8514497712254524,
      "step": 330
    },
    {
      "epoch": 0.6525444363282201,
      "grad_norm": 0.4563934748637001,
      "learning_rate": 8.520459688527091e-06,
      "loss": 0.5108,
      "mean_token_accuracy": 0.849525935947895,
      "step": 335
    },
    {
      "epoch": 0.6622839055271488,
      "grad_norm": 0.4756063591717307,
      "learning_rate": 8.459528307147066e-06,
      "loss": 0.51,
      "mean_token_accuracy": 0.8501726359128952,
      "step": 340
    },
    {
      "epoch": 0.6720233747260774,
      "grad_norm": 0.5732019741491505,
      "learning_rate": 8.397594983359591e-06,
      "loss": 0.5062,
      "mean_token_accuracy": 0.850397090613842,
      "step": 345
    },
    {
      "epoch": 0.6817628439250061,
      "grad_norm": 0.5267653966986828,
      "learning_rate": 8.334677654183254e-06,
      "loss": 0.5065,
      "mean_token_accuracy": 0.8505988359451294,
      "step": 350
    },
    {
      "epoch": 0.6915023131239347,
      "grad_norm": 0.4796149747636043,
      "learning_rate": 8.27079454162252e-06,
      "loss": 0.5028,
      "mean_token_accuracy": 0.851087860763073,
      "step": 355
    },
    {
      "epoch": 0.7012417823228634,
      "grad_norm": 0.4809967856664383,
      "learning_rate": 8.205964147390313e-06,
      "loss": 0.5084,
      "mean_token_accuracy": 0.8496400877833367,
      "step": 360
    },
    {
      "epoch": 0.710981251521792,
      "grad_norm": 0.4348534092140185,
      "learning_rate": 8.140205247549583e-06,
      "loss": 0.4983,
      "mean_token_accuracy": 0.8522587567567825,
      "step": 365
    },
    {
      "epoch": 0.7207207207207207,
      "grad_norm": 0.4799630529257426,
      "learning_rate": 8.073536887075417e-06,
      "loss": 0.5119,
      "mean_token_accuracy": 0.8493492469191551,
      "step": 370
    },
    {
      "epoch": 0.7304601899196493,
      "grad_norm": 0.5305052950944196,
      "learning_rate": 8.005978374339264e-06,
      "loss": 0.4946,
      "mean_token_accuracy": 0.8531364649534225,
      "step": 375
    },
    {
      "epoch": 0.740199659118578,
      "grad_norm": 0.46401531374429855,
      "learning_rate": 7.937549275516882e-06,
      "loss": 0.493,
      "mean_token_accuracy": 0.8535278528928757,
      "step": 380
    },
    {
      "epoch": 0.7499391283175066,
      "grad_norm": 0.46753323308197925,
      "learning_rate": 7.868269408921614e-06,
      "loss": 0.504,
      "mean_token_accuracy": 0.8511819407343865,
      "step": 385
    },
    {
      "epoch": 0.7596785975164354,
      "grad_norm": 0.43897436674183826,
      "learning_rate": 7.798158839264645e-06,
      "loss": 0.4983,
      "mean_token_accuracy": 0.8521765768527985,
      "step": 390
    },
    {
      "epoch": 0.7694180667153641,
      "grad_norm": 0.4030095081507055,
      "learning_rate": 7.7272378718439e-06,
      "loss": 0.5092,
      "mean_token_accuracy": 0.8500049978494644,
      "step": 395
    },
    {
      "epoch": 0.7791575359142927,
      "grad_norm": 0.393465790048579,
      "learning_rate": 7.655527046663254e-06,
      "loss": 0.5022,
      "mean_token_accuracy": 0.8510702222585678,
      "step": 400
    },
    {
      "epoch": 0.7888970051132214,
      "grad_norm": 0.49269595855244275,
      "learning_rate": 7.5830471324837765e-06,
      "loss": 0.4945,
      "mean_token_accuracy": 0.8529314771294594,
      "step": 405
    },
    {
      "epoch": 0.79863647431215,
      "grad_norm": 0.4729001525106615,
      "learning_rate": 7.5098191208087144e-06,
      "loss": 0.5,
      "mean_token_accuracy": 0.8525989070534706,
      "step": 410
    },
    {
      "epoch": 0.8083759435110787,
      "grad_norm": 0.44720074409020516,
      "learning_rate": 7.4358642198039835e-06,
      "loss": 0.4946,
      "mean_token_accuracy": 0.8530281245708465,
      "step": 415
    },
    {
      "epoch": 0.8181154127100073,
      "grad_norm": 0.4192773521625985,
      "learning_rate": 7.36120384815588e-06,
      "loss": 0.4927,
      "mean_token_accuracy": 0.8539465010166168,
      "step": 420
    },
    {
      "epoch": 0.827854881908936,
      "grad_norm": 0.41037683317185314,
      "learning_rate": 7.285859628867851e-06,
      "loss": 0.4952,
      "mean_token_accuracy": 0.8532393842935562,
      "step": 425
    },
    {
      "epoch": 0.8375943511078646,
      "grad_norm": 0.4536782151813187,
      "learning_rate": 7.209853382998077e-06,
      "loss": 0.4983,
      "mean_token_accuracy": 0.8528945103287697,
      "step": 430
    },
    {
      "epoch": 0.8473338203067933,
      "grad_norm": 0.4895833290473578,
      "learning_rate": 7.133207123339689e-06,
      "loss": 0.4939,
      "mean_token_accuracy": 0.853129243850708,
      "step": 435
    },
    {
      "epoch": 0.8570732895057219,
      "grad_norm": 0.4771689532642086,
      "learning_rate": 7.055943048045476e-06,
      "loss": 0.5002,
      "mean_token_accuracy": 0.8518661975860595,
      "step": 440
    },
    {
      "epoch": 0.8668127587046506,
      "grad_norm": 0.4439177979495324,
      "learning_rate": 6.978083534198878e-06,
      "loss": 0.4977,
      "mean_token_accuracy": 0.8526063248515129,
      "step": 445
    },
    {
      "epoch": 0.8765522279035792,
      "grad_norm": 0.45795986197220584,
      "learning_rate": 6.899651131333194e-06,
      "loss": 0.4876,
      "mean_token_accuracy": 0.8546572834253311,
      "step": 450
    },
    {
      "epoch": 0.8862916971025079,
      "grad_norm": 0.38084088770564956,
      "learning_rate": 6.82066855490081e-06,
      "loss": 0.5009,
      "mean_token_accuracy": 0.851148933172226,
      "step": 455
    },
    {
      "epoch": 0.8960311663014365,
      "grad_norm": 0.4552737931829716,
      "learning_rate": 6.741158679694403e-06,
      "loss": 0.4968,
      "mean_token_accuracy": 0.8524380102753639,
      "step": 460
    },
    {
      "epoch": 0.9057706355003652,
      "grad_norm": 0.38831398930391275,
      "learning_rate": 6.661144533221974e-06,
      "loss": 0.4897,
      "mean_token_accuracy": 0.8537176489830017,
      "step": 465
    },
    {
      "epoch": 0.9155101046992938,
      "grad_norm": 0.42794585960634596,
      "learning_rate": 6.58064928903767e-06,
      "loss": 0.4942,
      "mean_token_accuracy": 0.8529506504535675,
      "step": 470
    },
    {
      "epoch": 0.9252495738982226,
      "grad_norm": 0.42890975667323755,
      "learning_rate": 6.499696260030297e-06,
      "loss": 0.5064,
      "mean_token_accuracy": 0.8502799227833748,
      "step": 475
    },
    {
      "epoch": 0.9349890430971513,
      "grad_norm": 0.4478332747744812,
      "learning_rate": 6.418308891671484e-06,
      "loss": 0.4855,
      "mean_token_accuracy": 0.8555838361382484,
      "step": 480
    },
    {
      "epoch": 0.9447285122960799,
      "grad_norm": 0.435211802978247,
      "learning_rate": 6.336510755225447e-06,
      "loss": 0.4835,
      "mean_token_accuracy": 0.8558964654803276,
      "step": 485
    },
    {
      "epoch": 0.9544679814950086,
      "grad_norm": 0.4462726381437146,
      "learning_rate": 6.25432554092232e-06,
      "loss": 0.4898,
      "mean_token_accuracy": 0.8544702440500259,
      "step": 490
    },
    {
      "epoch": 0.9642074506939372,
      "grad_norm": 0.44503915866443544,
      "learning_rate": 6.171777051097037e-06,
      "loss": 0.4858,
      "mean_token_accuracy": 0.8554443955421448,
      "step": 495
    },
    {
      "epoch": 0.9739469198928659,
      "grad_norm": 0.47966098493432496,
      "learning_rate": 6.088889193295738e-06,
      "loss": 0.4929,
      "mean_token_accuracy": 0.8535514727234841,
      "step": 500
    },
    {
      "epoch": 0.9836863890917945,
      "grad_norm": 0.5437585793823457,
      "learning_rate": 6.005685973351708e-06,
      "loss": 0.4931,
      "mean_token_accuracy": 0.8531181156635285,
      "step": 505
    },
    {
      "epoch": 0.9934258582907232,
      "grad_norm": 0.4124694681433172,
      "learning_rate": 5.922191488432857e-06,
      "loss": 0.4917,
      "mean_token_accuracy": 0.8535135626792908,
      "step": 510
    },
    {
      "epoch": 1.0019478938397857,
      "grad_norm": 0.803029283965446,
      "learning_rate": 5.838429920062734e-06,
      "loss": 0.4727,
      "mean_token_accuracy": 0.8567549926894051,
      "step": 515
    },
    {
      "epoch": 1.0116873630387144,
      "grad_norm": 0.3927635667112885,
      "learning_rate": 5.754425527117118e-06,
      "loss": 0.4479,
      "mean_token_accuracy": 0.864441742002964,
      "step": 520
    },
    {
      "epoch": 1.021426832237643,
      "grad_norm": 0.45928387456979786,
      "learning_rate": 5.670202638798213e-06,
      "loss": 0.4598,
      "mean_token_accuracy": 0.8615871027112008,
      "step": 525
    },
    {
      "epoch": 1.0311663014365717,
      "grad_norm": 0.517389580126703,
      "learning_rate": 5.585785647588458e-06,
      "loss": 0.4572,
      "mean_token_accuracy": 0.8620010375976562,
      "step": 530
    },
    {
      "epoch": 1.0409057706355003,
      "grad_norm": 0.37886924917283415,
      "learning_rate": 5.501199002186024e-06,
      "loss": 0.455,
      "mean_token_accuracy": 0.8625968441367149,
      "step": 535
    },
    {
      "epoch": 1.050645239834429,
      "grad_norm": 0.39773153996958316,
      "learning_rate": 5.416467200424032e-06,
      "loss": 0.45,
      "mean_token_accuracy": 0.8637859463691712,
      "step": 540
    },
    {
      "epoch": 1.0603847090333576,
      "grad_norm": 0.40508395528042257,
      "learning_rate": 5.33161478217552e-06,
      "loss": 0.4516,
      "mean_token_accuracy": 0.8634377360343933,
      "step": 545
    },
    {
      "epoch": 1.0701241782322863,
      "grad_norm": 0.37429603209812456,
      "learning_rate": 5.246666322246267e-06,
      "loss": 0.4445,
      "mean_token_accuracy": 0.8651037693023682,
      "step": 550
    },
    {
      "epoch": 1.079863647431215,
      "grad_norm": 0.47125189912053295,
      "learning_rate": 5.1616464232574635e-06,
      "loss": 0.4626,
      "mean_token_accuracy": 0.8602706581354141,
      "step": 555
    },
    {
      "epoch": 1.0896031166301436,
      "grad_norm": 0.435383825468146,
      "learning_rate": 5.076579708520355e-06,
      "loss": 0.4497,
      "mean_token_accuracy": 0.8635805040597916,
      "step": 560
    },
    {
      "epoch": 1.0993425858290724,
      "grad_norm": 0.412421290706311,
      "learning_rate": 4.991490814904888e-06,
      "loss": 0.4378,
      "mean_token_accuracy": 0.8668369174003601,
      "step": 565
    },
    {
      "epoch": 1.109082055028001,
      "grad_norm": 0.41209358309681204,
      "learning_rate": 4.906404385704402e-06,
      "loss": 0.4525,
      "mean_token_accuracy": 0.8637306377291679,
      "step": 570
    },
    {
      "epoch": 1.1188215242269297,
      "grad_norm": 0.3821281290056491,
      "learning_rate": 4.82134506349851e-06,
      "loss": 0.4564,
      "mean_token_accuracy": 0.8625956058502198,
      "step": 575
    },
    {
      "epoch": 1.1285609934258582,
      "grad_norm": 0.34980242368851916,
      "learning_rate": 4.736337483016138e-06,
      "loss": 0.4513,
      "mean_token_accuracy": 0.8634121060371399,
      "step": 580
    },
    {
      "epoch": 1.138300462624787,
      "grad_norm": 0.39297932611221925,
      "learning_rate": 4.651406264000871e-06,
      "loss": 0.4512,
      "mean_token_accuracy": 0.8632175624370575,
      "step": 585
    },
    {
      "epoch": 1.1480399318237156,
      "grad_norm": 0.4317512663581136,
      "learning_rate": 4.5665760040806174e-06,
      "loss": 0.4558,
      "mean_token_accuracy": 0.8621700823307037,
      "step": 590
    },
    {
      "epoch": 1.1577794010226443,
      "grad_norm": 0.4166519413744608,
      "learning_rate": 4.481871271643698e-06,
      "loss": 0.4543,
      "mean_token_accuracy": 0.8628953084349632,
      "step": 595
    },
    {
      "epoch": 1.1675188702215729,
      "grad_norm": 0.3817664896689048,
      "learning_rate": 4.397316598723385e-06,
      "loss": 0.4599,
      "mean_token_accuracy": 0.8614401906728745,
      "step": 600
    },
    {
      "epoch": 1.1772583394205016,
      "grad_norm": 0.39223028227054235,
      "learning_rate": 4.312936473892984e-06,
      "loss": 0.4559,
      "mean_token_accuracy": 0.8621881052851676,
      "step": 605
    },
    {
      "epoch": 1.1869978086194302,
      "grad_norm": 0.39870192645434044,
      "learning_rate": 4.228755335173488e-06,
      "loss": 0.4554,
      "mean_token_accuracy": 0.8622724115848541,
      "step": 610
    },
    {
      "epoch": 1.196737277818359,
      "grad_norm": 0.3968827579618214,
      "learning_rate": 4.1447975629559e-06,
      "loss": 0.4496,
      "mean_token_accuracy": 0.8639340966939926,
      "step": 615
    },
    {
      "epoch": 1.2064767470172875,
      "grad_norm": 0.3525974840125754,
      "learning_rate": 4.061087472940204e-06,
      "loss": 0.4468,
      "mean_token_accuracy": 0.8643500834703446,
      "step": 620
    },
    {
      "epoch": 1.2162162162162162,
      "grad_norm": 0.34784808537613615,
      "learning_rate": 3.977649309093113e-06,
      "loss": 0.4463,
      "mean_token_accuracy": 0.8645710095763206,
      "step": 625
    },
    {
      "epoch": 1.225955685415145,
      "grad_norm": 0.3588527258157974,
      "learning_rate": 3.89450723662657e-06,
      "loss": 0.4517,
      "mean_token_accuracy": 0.863483439385891,
      "step": 630
    },
    {
      "epoch": 1.2356951546140735,
      "grad_norm": 0.495886348965792,
      "learning_rate": 3.8116853349990574e-06,
      "loss": 0.4507,
      "mean_token_accuracy": 0.8636451244354248,
      "step": 635
    },
    {
      "epoch": 1.245434623813002,
      "grad_norm": 0.3842625748309975,
      "learning_rate": 3.729207590941753e-06,
      "loss": 0.4377,
      "mean_token_accuracy": 0.8670691177248955,
      "step": 640
    },
    {
      "epoch": 1.2551740930119308,
      "grad_norm": 0.4426936471853252,
      "learning_rate": 3.647097891511536e-06,
      "loss": 0.4495,
      "mean_token_accuracy": 0.8640359625220299,
      "step": 645
    },
    {
      "epoch": 1.2649135622108596,
      "grad_norm": 0.3707914127374137,
      "learning_rate": 3.565380017172854e-06,
      "loss": 0.4397,
      "mean_token_accuracy": 0.8666136890649796,
      "step": 650
    },
    {
      "epoch": 1.2746530314097881,
      "grad_norm": 0.35267532874755536,
      "learning_rate": 3.4840776349104755e-06,
      "loss": 0.4539,
      "mean_token_accuracy": 0.8626947477459908,
      "step": 655
    },
    {
      "epoch": 1.284392500608717,
      "grad_norm": 0.3749743227529099,
      "learning_rate": 3.4032142913750956e-06,
      "loss": 0.4497,
      "mean_token_accuracy": 0.8637418314814568,
      "step": 660
    },
    {
      "epoch": 1.2941319698076454,
      "grad_norm": 0.37600006002277636,
      "learning_rate": 3.322813406063794e-06,
      "loss": 0.4559,
      "mean_token_accuracy": 0.8622224271297455,
      "step": 665
    },
    {
      "epoch": 1.3038714390065742,
      "grad_norm": 0.4005804398308692,
      "learning_rate": 3.242898264537331e-06,
      "loss": 0.4521,
      "mean_token_accuracy": 0.8632362619042396,
      "step": 670
    },
    {
      "epoch": 1.3136109082055027,
      "grad_norm": 0.38630202996272256,
      "learning_rate": 3.1634920116762175e-06,
      "loss": 0.4499,
      "mean_token_accuracy": 0.8635998621582985,
      "step": 675
    },
    {
      "epoch": 1.3233503774044315,
      "grad_norm": 0.36180685239531096,
      "learning_rate": 3.0846176449775363e-06,
      "loss": 0.4508,
      "mean_token_accuracy": 0.8636634424328804,
      "step": 680
    },
    {
      "epoch": 1.33308984660336,
      "grad_norm": 0.3357931590954893,
      "learning_rate": 3.0062980078944515e-06,
      "loss": 0.4379,
      "mean_token_accuracy": 0.8665053129196167,
      "step": 685
    },
    {
      "epoch": 1.3428293158022888,
      "grad_norm": 0.35662293115272325,
      "learning_rate": 2.9285557832203328e-06,
      "loss": 0.4458,
      "mean_token_accuracy": 0.8648849859833717,
      "step": 690
    },
    {
      "epoch": 1.3525687850012174,
      "grad_norm": 0.35254913645932934,
      "learning_rate": 2.851413486519388e-06,
      "loss": 0.4413,
      "mean_token_accuracy": 0.8654858738183975,
      "step": 695
    },
    {
      "epoch": 1.3623082542001461,
      "grad_norm": 0.3448432687225989,
      "learning_rate": 2.774893459605766e-06,
      "loss": 0.4431,
      "mean_token_accuracy": 0.86555365473032,
      "step": 700
    },
    {
      "epoch": 1.3720477233990747,
      "grad_norm": 0.35405617501926967,
      "learning_rate": 2.69901786407295e-06,
      "loss": 0.444,
      "mean_token_accuracy": 0.8651208564639091,
      "step": 705
    },
    {
      "epoch": 1.3817871925980034,
      "grad_norm": 0.36859590132072323,
      "learning_rate": 2.6238086748753587e-06,
      "loss": 0.456,
      "mean_token_accuracy": 0.8627100110054016,
      "step": 710
    },
    {
      "epoch": 1.3915266617969322,
      "grad_norm": 0.35246470987862855,
      "learning_rate": 2.5492876739639912e-06,
      "loss": 0.4533,
      "mean_token_accuracy": 0.8628792524337768,
      "step": 715
    },
    {
      "epoch": 1.4012661309958607,
      "grad_norm": 0.36351033252725057,
      "learning_rate": 2.475476443977996e-06,
      "loss": 0.4469,
      "mean_token_accuracy": 0.8642540082335473,
      "step": 720
    },
    {
      "epoch": 1.4110056001947893,
      "grad_norm": 0.35938233389923585,
      "learning_rate": 2.40239636199393e-06,
      "loss": 0.4451,
      "mean_token_accuracy": 0.8650069192051888,
      "step": 725
    },
    {
      "epoch": 1.420745069393718,
      "grad_norm": 0.3673326064574297,
      "learning_rate": 2.3300685933345656e-06,
      "loss": 0.4485,
      "mean_token_accuracy": 0.8642319470643998,
      "step": 730
    },
    {
      "epoch": 1.4304845385926468,
      "grad_norm": 0.37166399663475924,
      "learning_rate": 2.2585140854390432e-06,
      "loss": 0.4496,
      "mean_token_accuracy": 0.8641144469380379,
      "step": 735
    },
    {
      "epoch": 1.4402240077915753,
      "grad_norm": 0.34969080725493995,
      "learning_rate": 2.187753561796097e-06,
      "loss": 0.449,
      "mean_token_accuracy": 0.8638374775648117,
      "step": 740
    },
    {
      "epoch": 1.449963476990504,
      "grad_norm": 0.33346605175419364,
      "learning_rate": 2.117807515942163e-06,
      "loss": 0.4487,
      "mean_token_accuracy": 0.8639461770653725,
      "step": 745
    },
    {
      "epoch": 1.4597029461894326,
      "grad_norm": 0.3402282068550279,
      "learning_rate": 2.0486962055260744e-06,
      "loss": 0.4381,
      "mean_token_accuracy": 0.8668898791074753,
      "step": 750
    },
    {
      "epoch": 1.4694424153883614,
      "grad_norm": 0.3454615445905308,
      "learning_rate": 1.9804396464420798e-06,
      "loss": 0.4407,
      "mean_token_accuracy": 0.8662415385246277,
      "step": 755
    },
    {
      "epoch": 1.47918188458729,
      "grad_norm": 0.3815154168732918,
      "learning_rate": 1.9130576070328695e-06,
      "loss": 0.4502,
      "mean_token_accuracy": 0.863922019302845,
      "step": 760
    },
    {
      "epoch": 1.4889213537862187,
      "grad_norm": 0.35782414062256107,
      "learning_rate": 1.8465696023643115e-06,
      "loss": 0.4484,
      "mean_token_accuracy": 0.8640252217650414,
      "step": 765
    },
    {
      "epoch": 1.4986608229851472,
      "grad_norm": 0.34966179386580715,
      "learning_rate": 1.7809948885735295e-06,
      "loss": 0.4476,
      "mean_token_accuracy": 0.8640663206577301,
      "step": 770
    },
    {
      "epoch": 1.508400292184076,
      "grad_norm": 0.3283826370466805,
      "learning_rate": 1.7163524572919748e-06,
      "loss": 0.4535,
      "mean_token_accuracy": 0.8629004299640656,
      "step": 775
    },
    {
      "epoch": 1.5181397613830048,
      "grad_norm": 0.3309412101835561,
      "learning_rate": 1.6526610301451028e-06,
      "loss": 0.4374,
      "mean_token_accuracy": 0.8666589662432671,
      "step": 780
    },
    {
      "epoch": 1.5278792305819333,
      "grad_norm": 0.35820157894670385,
      "learning_rate": 1.5899390533302538e-06,
      "loss": 0.4387,
      "mean_token_accuracy": 0.8667290091514588,
      "step": 785
    },
    {
      "epoch": 1.5376186997808619,
      "grad_norm": 0.34472450626785944,
      "learning_rate": 1.5282046922742876e-06,
      "loss": 0.4502,
      "mean_token_accuracy": 0.8635128363966942,
      "step": 790
    },
    {
      "epoch": 1.5473581689797906,
      "grad_norm": 0.31690650065908227,
      "learning_rate": 1.4674758263725614e-06,
      "loss": 0.4461,
      "mean_token_accuracy": 0.8644041374325753,
      "step": 795
    },
    {
      "epoch": 1.5570976381787194,
      "grad_norm": 0.3261992242400567,
      "learning_rate": 1.4077700438107183e-06,
      "loss": 0.445,
      "mean_token_accuracy": 0.8651425749063492,
      "step": 800
    },
    {
      "epoch": 1.566837107377648,
      "grad_norm": 0.37132210271715693,
      "learning_rate": 1.3491046364708294e-06,
      "loss": 0.445,
      "mean_token_accuracy": 0.8648254871368408,
      "step": 805
    },
    {
      "epoch": 1.5765765765765765,
      "grad_norm": 0.34739884194127096,
      "learning_rate": 1.2914965949233572e-06,
      "loss": 0.4474,
      "mean_token_accuracy": 0.8643729150295257,
      "step": 810
    },
    {
      "epoch": 1.5863160457755052,
      "grad_norm": 0.3092847696386134,
      "learning_rate": 1.2349626035063705e-06,
      "loss": 0.4389,
      "mean_token_accuracy": 0.8666358023881913,
      "step": 815
    },
    {
      "epoch": 1.596055514974434,
      "grad_norm": 0.33368514787450476,
      "learning_rate": 1.1795190354934587e-06,
      "loss": 0.4635,
      "mean_token_accuracy": 0.8606240957975387,
      "step": 820
    },
    {
      "epoch": 1.6057949841733625,
      "grad_norm": 0.3355701607651583,
      "learning_rate": 1.1251819483517334e-06,
      "loss": 0.4469,
      "mean_token_accuracy": 0.8647612199187279,
      "step": 825
    },
    {
      "epoch": 1.615534453372291,
      "grad_norm": 0.3215469441235877,
      "learning_rate": 1.0719670790912928e-06,
      "loss": 0.4479,
      "mean_token_accuracy": 0.8641064539551735,
      "step": 830
    },
    {
      "epoch": 1.6252739225712198,
      "grad_norm": 0.3331920310980409,
      "learning_rate": 1.019889839707498e-06,
      "loss": 0.447,
      "mean_token_accuracy": 0.8645351231098175,
      "step": 835
    },
    {
      "epoch": 1.6350133917701486,
      "grad_norm": 0.3311460501397384,
      "learning_rate": 9.689653127173743e-07,
      "loss": 0.4548,
      "mean_token_accuracy": 0.8624306350946427,
      "step": 840
    },
    {
      "epoch": 1.6447528609690771,
      "grad_norm": 0.3387701231769685,
      "learning_rate": 9.192082467914465e-07,
      "loss": 0.444,
      "mean_token_accuracy": 0.8649628892540931,
      "step": 845
    },
    {
      "epoch": 1.654492330168006,
      "grad_norm": 0.33969198086970775,
      "learning_rate": 8.706330524822548e-07,
      "loss": 0.4413,
      "mean_token_accuracy": 0.865912164747715,
      "step": 850
    },
    {
      "epoch": 1.6642317993669344,
      "grad_norm": 0.32626435288941413,
      "learning_rate": 8.232537980507848e-07,
      "loss": 0.4454,
      "mean_token_accuracy": 0.8650734156370163,
      "step": 855
    },
    {
      "epoch": 1.6739712685658632,
      "grad_norm": 0.32534723430048623,
      "learning_rate": 7.770842053920585e-07,
      "loss": 0.4424,
      "mean_token_accuracy": 0.8653772249817848,
      "step": 860
    },
    {
      "epoch": 1.683710737764792,
      "grad_norm": 0.33973250353406415,
      "learning_rate": 7.321376460610136e-07,
      "loss": 0.4398,
      "mean_token_accuracy": 0.865781269967556,
      "step": 865
    },
    {
      "epoch": 1.6934502069637205,
      "grad_norm": 0.31477206750092435,
      "learning_rate": 6.884271373998608e-07,
      "loss": 0.4402,
      "mean_token_accuracy": 0.8660067468881607,
      "step": 870
    },
    {
      "epoch": 1.703189676162649,
      "grad_norm": 0.3537664329765484,
      "learning_rate": 6.459653387680248e-07,
      "loss": 0.4426,
      "mean_token_accuracy": 0.8656087100505829,
      "step": 875
    },
    {
      "epoch": 1.7129291453615778,
      "grad_norm": 0.32858889216418335,
      "learning_rate": 6.047645478757635e-07,
      "loss": 0.4367,
      "mean_token_accuracy": 0.8670719146728516,
      "step": 880
    },
    {
      "epoch": 1.7226686145605066,
      "grad_norm": 0.32864732265316704,
      "learning_rate": 5.648366972225222e-07,
      "loss": 0.4527,
      "mean_token_accuracy": 0.8630405649542808,
      "step": 885
    },
    {
      "epoch": 1.7324080837594351,
      "grad_norm": 0.3173505795175411,
      "learning_rate": 5.261933506410722e-07,
      "loss": 0.4401,
      "mean_token_accuracy": 0.8665132194757461,
      "step": 890
    },
    {
      "epoch": 1.7421475529583637,
      "grad_norm": 0.36040952773775503,
      "learning_rate": 4.888456999484098e-07,
      "loss": 0.4465,
      "mean_token_accuracy": 0.8646954327821732,
      "step": 895
    },
    {
      "epoch": 1.7518870221572924,
      "grad_norm": 0.31604383070212816,
      "learning_rate": 4.528045617044019e-07,
      "loss": 0.443,
      "mean_token_accuracy": 0.8652036920189857,
      "step": 900
    },
    {
      "epoch": 1.7616264913562212,
      "grad_norm": 0.31389323834734445,
      "learning_rate": 4.180803740791156e-07,
      "loss": 0.4426,
      "mean_token_accuracy": 0.8656812936067582,
      "step": 905
    },
    {
      "epoch": 1.7713659605551497,
      "grad_norm": 0.3097110381509705,
      "learning_rate": 3.846831938297324e-07,
      "loss": 0.4468,
      "mean_token_accuracy": 0.8643257409334183,
      "step": 910
    },
    {
      "epoch": 1.7811054297540783,
      "grad_norm": 0.31179540512244613,
      "learning_rate": 3.5262269338792623e-07,
      "loss": 0.4447,
      "mean_token_accuracy": 0.8651932507753373,
      "step": 915
    },
    {
      "epoch": 1.790844898953007,
      "grad_norm": 0.33502042463366694,
      "learning_rate": 3.219081580585548e-07,
      "loss": 0.4508,
      "mean_token_accuracy": 0.863429008424282,
      "step": 920
    },
    {
      "epoch": 1.8005843681519358,
      "grad_norm": 0.3164739894198197,
      "learning_rate": 2.9254848333046817e-07,
      "loss": 0.4528,
      "mean_token_accuracy": 0.8630515649914742,
      "step": 925
    },
    {
      "epoch": 1.8103238373508643,
      "grad_norm": 0.3030075688047627,
      "learning_rate": 2.645521723002037e-07,
      "loss": 0.4507,
      "mean_token_accuracy": 0.8635053560137749,
      "step": 930
    },
    {
      "epoch": 1.8200633065497929,
      "grad_norm": 0.3371646202292577,
      "learning_rate": 2.3792733320934348e-07,
      "loss": 0.4441,
      "mean_token_accuracy": 0.8654543533921242,
      "step": 935
    },
    {
      "epoch": 1.8298027757487216,
      "grad_norm": 0.3074575906934633,
      "learning_rate": 2.12681677096217e-07,
      "loss": 0.4351,
      "mean_token_accuracy": 0.8675669968128205,
      "step": 940
    },
    {
      "epoch": 1.8395422449476504,
      "grad_norm": 0.30631898093344184,
      "learning_rate": 1.888225155626433e-07,
      "loss": 0.444,
      "mean_token_accuracy": 0.8653656959533691,
      "step": 945
    },
    {
      "epoch": 1.8492817141465792,
      "grad_norm": 0.2958594503607962,
      "learning_rate": 1.6635675865635859e-07,
      "loss": 0.4568,
      "mean_token_accuracy": 0.8619549512863159,
      "step": 950
    },
    {
      "epoch": 1.8590211833455077,
      "grad_norm": 0.31258753396965294,
      "learning_rate": 1.4529091286973994e-07,
      "loss": 0.444,
      "mean_token_accuracy": 0.8656645834445953,
      "step": 955
    },
    {
      "epoch": 1.8687606525444362,
      "grad_norm": 0.30999125345039114,
      "learning_rate": 1.2563107925540774e-07,
      "loss": 0.4444,
      "mean_token_accuracy": 0.8654705569148063,
      "step": 960
    },
    {
      "epoch": 1.878500121743365,
      "grad_norm": 0.3160340098992061,
      "learning_rate": 1.0738295165924783e-07,
      "loss": 0.4459,
      "mean_token_accuracy": 0.8648449763655662,
      "step": 965
    },
    {
      "epoch": 1.8882395909422938,
      "grad_norm": 0.3187758679651324,
      "learning_rate": 9.055181507137245e-08,
      "loss": 0.444,
      "mean_token_accuracy": 0.8654528453946113,
      "step": 970
    },
    {
      "epoch": 1.8979790601412223,
      "grad_norm": 0.33617665839091787,
      "learning_rate": 7.514254409549005e-08,
      "loss": 0.4481,
      "mean_token_accuracy": 0.8641701668500901,
      "step": 975
    },
    {
      "epoch": 1.9077185293401508,
      "grad_norm": 0.31232672260705024,
      "learning_rate": 6.115960153712963e-08,
      "loss": 0.4414,
      "mean_token_accuracy": 0.8660656422376632,
      "step": 980
    },
    {
      "epoch": 1.9174579985390796,
      "grad_norm": 0.303119007372895,
      "learning_rate": 4.860703711113246e-08,
      "loss": 0.4362,
      "mean_token_accuracy": 0.8676238685846329,
      "step": 985
    },
    {
      "epoch": 1.9271974677380084,
      "grad_norm": 0.30341116951092534,
      "learning_rate": 3.748848626878132e-08,
      "loss": 0.4508,
      "mean_token_accuracy": 0.8636912703514099,
      "step": 990
    },
    {
      "epoch": 1.936936936936937,
      "grad_norm": 0.31515978754450974,
      "learning_rate": 2.7807169144906108e-08,
      "loss": 0.4528,
      "mean_token_accuracy": 0.8630853027105332,
      "step": 995
    },
    {
      "epoch": 1.9466764061358655,
      "grad_norm": 0.30748596132213357,
      "learning_rate": 1.9565889625275945e-08,
      "loss": 0.4558,
      "mean_token_accuracy": 0.8628044292330742,
      "step": 1000
    },
    {
      "epoch": 1.9564158753347942,
      "grad_norm": 0.30248992557073057,
      "learning_rate": 1.2767034534540978e-08,
      "loss": 0.4467,
      "mean_token_accuracy": 0.86439578384161,
      "step": 1005
    },
    {
      "epoch": 1.966155344533723,
      "grad_norm": 0.3256862744869616,
      "learning_rate": 7.412572944965335e-09,
      "loss": 0.4548,
      "mean_token_accuracy": 0.8625592529773712,
      "step": 1010
    },
    {
      "epoch": 1.9758948137326515,
      "grad_norm": 0.3147340514506444,
      "learning_rate": 3.5040556061483043e-09,
      "loss": 0.4422,
      "mean_token_accuracy": 0.8657747611403466,
      "step": 1015
    },
    {
      "epoch": 1.98563428293158,
      "grad_norm": 0.32764257560082577,
      "learning_rate": 1.0426144958985974e-09,
      "loss": 0.4419,
      "mean_token_accuracy": 0.8660721600055694,
      "step": 1020
    },
    {
      "epoch": 1.9953737521305088,
      "grad_norm": 0.3098108021812636,
      "learning_rate": 2.8962492393258546e-11,
      "loss": 0.4452,
      "mean_token_accuracy": 0.8650736406445503,
      "step": 1025
    },
    {
      "epoch": 1.9973216459702945,
      "mean_token_accuracy": 0.8681519404053688,
      "step": 1026,
      "total_flos": 1074983740637184.0,
      "train_loss": 0.49954891321022377,
      "train_runtime": 168401.8035,
      "train_samples_per_second": 0.78,
      "train_steps_per_second": 0.006
    }
  ],
  "logging_steps": 5,
  "max_steps": 1026,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1074983740637184.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}